THE ROLE OF PROLACTIN IN MAMMOGENESIS AND LACTOGENESIS
Laurence S. Jacobs Washington University School of Medicine Department of Medicine-Metabolism 660 S. Euclid St. Louis, Missouri 63110 OUTLINE I. II.
Introduction Control of Prolactin Secretion Hypothalamic influences Estrogen Thyroid function Drugs
III.
Regulation of Mammary Growth and Development Normal hormonal requirements Insights gained from disease states Effects of pregnancy
IV.
Regulation of Milk Production Determinants of postpartum lactation Relationship between serum PRL levels and lactation Milk PRL levels PRL receptor affinity Interactions with steroids I.
INTRODUCTION
Although a great deal of new information about the physiologic and pathologic regulation of prolactin secretion in man has become available in recent years, precise delineation of the role of prolactin in human mammogenesis and lactogenesis has been difficult. This should not be surprising in light of the fact that its role in these processes is imperfectly understood in sub-primate mammals.
173 H.-D. Dellmann et al. (eds.), Comparative Endocrinology of Prolactin © Plenum Press, New York 1977
L.S, JACOBS
174
Though detailed information has been accumulated 011 hormonal requirements for the maintenance and function of mammary explants in short-term tissue culture in vitro, the relevance of these observations to in vivo lactational physiology remains unclear. A substantial body of literature dealing with lactation and its vascular cytologic, biochemical, and physiologic aspects has been developed during the past 30 years, but many aspects of breast function remain poorly understood (Larson and Smith, 1974). At the outset, it should be noted that many reasons for the relative paucity of our understanding of breast physiology are apparent. One of the most important of these is the large number of hormonal signals to which the breast appears to be responsive. Estrogen, progesterone, placental lactogen, and pituitary prolactin play important roles in determining breast size and in modulating breast function. Additionally, there is substantial albeit less compelling existence that lactational performance is in part dependent on normal thyroid and adrenal function; softer evidence further indicates the likelihood that the mammary gland requires insulin for normal functioning. The influence of these latter factors on normal breast development remains essentially unknown. We cannot extrapolate with very much certainty from the rodent models which have served as the basis for much of the in vitro information to the human lactational situation. Because of the dominant inhibitory hypothalamic control over prolactin secretion, lesions of the stalk or hypothalamus in man which do not also totally destroy the lactotrope cell population may cause marked increases in prolactin secretion (Foley, Jacobs,
Table 1.
HORMONAL FACTORS IN HUMAN MAMMARY GROWTH AND DEVELOPMENT I.
II.
III.
IV.
Hormones of Established Prime Importance Estrogens Progesterone Prolactin Placental Lactogen (chorionic somatomammotropin) Hormones Probably Required in a Permissive Way Thyroxine/Triiodothyronine Insulin Cortisol Hormones Implicated by Animal Studies but Probably not Relevant to Human Mammogenesis Growth Hormone Aldosterone Hormones Inhibiting Breast Development Androgens
ROLE IN MAMMOGENESIS AND LACTOGENESIS
175
Hoffman, Daughaday, and Blizzard, 1972; Snyder, Jacobs, Rabello, Sterling, Shore, Utiger, and Daughaday, 1974). One consequence of this fact is that it is hazardous to assume ablation of PRL 1 secretion following lesions of the hypothalamic-pituitary axis. Rather, empirical validation of PRL deficiency, based on radioimmunoassay measurement of circulating hormone levels, is required. Further, measurement of basal levels is insufficient. It should also be demonstrated that administration of a potent stimulus to PRL secretion such as TRH, which acts directly on the adenohypophysial lactotrope cell (MachI in , Jacobs, Cirulis, Kimes, and Miller, 1974), fails to elicit an increase in PRL secretion before one may safely conclude that hypolactotropism has been achieved. These considerations loom large when one considers the type of experimental evidence which would be required for the validation of an experimental model of PRL deficiency. An even greater paucity of useful information exists regarding the role of prolactin in the initiation and control of mammogenesis in man. The difficulties in advancing our understanding of this problem are related primarily to the lack of a suitable experimental model. Two features contribute: first, the scarcity and expense of a suitable primate species for study; second, the technical difficulties in producing long-term isolated deficiency of prolactin without concomitantly affecting the secretion of other pituitary hormones and their target organ secretions. Nonetheless, experimental evidence supporting a major role for prolactin in animal mammogenesis is compelling (Meites and Nicoll, 1966). In order to study the role of the hormone in mammogenesis, one should choose for study an animal in which mammary growth and development is readily identified as a normal maturational event. It would be preferable to establish a state of PRL deficiency in early postnatal life, and it would be essential of course that the PRL deficiency be an isolated one. These considerations make it difficult to draw conclusions regarding the physiologic role of prolactin in an animal, like some strains of mice, in whom growth hormone is well documented to have a potent mammotropic effect. This brief sketch of the formidable technical barriers which have thus far impeded more rapid progress in our understanding of breast maturation and function should serve as a background for the ensuing discussion of the control of prolactin secretion. We shall return later to the question of experimental approaches to the un-
lAbbreviations used in this manuscript are: PRL: prolactin; TRH: thyrotropin releasing hormone; PIF: prolactin inhibiting factor; TSH: thyrotropin; PRH: prolactin releasing hormone; GH: growth hormone.
L.S. JACOBS
176 derstanding of human breast maturation and function.
After even the briefest consideration of some of these issues it is not hard to understand why we have yet to see a definitive experimental study of the role of prolactin in in vivo mammogenesis. Total ablation of adenohypophysial endocrine functions, followed by fractional replacement therapy with thyroid hormone, glucocorticoids, sex steroids, and growth hormone, might shed some light on this problem, but would be a tedious and difficult longterm experiment. II.
CONTROL OF PROLACTIN SECRETION
Unlike the remainder of the anterior pituitary hormones in man, prolactin is regulated in vivo by a dominant inhibitory influence of the hypothalamus. Although it is clear from a large body of experimental evidence that dopaminergic mechanisms are involved in the inhibition of PRL secretion, the major locus at which dopamine may act in vivo is not clear. Considerable data in animal experiments as well as in man point to an effect of dopamine to suppress PRL secretion directly from the adenohypophysis (Birge, Jacobs, Hammer, and Daughaday, 1970; Macleod and Lehmeyer, 1974; Woolf, Jacobs, Donofrio, Burday, and Schalch, 1974; Shaar and Clemens, 1974). However, the hypothalamus is very rich in dopamine (Fuxe, 1974) and this compound is found in hypophysial portal vessel blood (Ben-Jonathan, Oliver, and Mical, 1976). It is conceivable that dopaminergic mechanisms may operate in vivo on both the hypothalamus and the pituitary. In the clinical arena, agents which are dopamine agonists, such as the precursor L-dopa or the ergot derivative 2-a-bromoergocryptine, have been used to suppress prolactin secretion and restore menses and fertility in hyperprolactinemic-amenorrheic women (del Pozo, Varga, Wyss, Tolis, Friesen, Wenner, Vetter, and Uettwiler, 1974). At present, it is not clear whether a prolactin inhibiting factor (PIF) of peptide nature exists as a hypothalamic entity separate from dopamine or not (Shaar and Clemens, 1974). In addition to the inhibitory effect which is readily unmasked
in vivo by experimental transection of the pituitary stalk, evidence
has accumulated that a quantitatively less important hypothalamic stimulatQry effect on PRL secretion may also exist. Data in support of this concept have come from studies on avian (Nicoll, Fiorindo, McKennee, and Parsons, 1970) and porcine (Valverde and Chieffo, 1971) hypothalamus. The biochemical interpretation most appropriate to these physiologic data renlain unclear, however. It has been established now for half a decade that TRH, a tripeptide of known hypothalamic residence, is a potent stimulator of PRL as well as of TSH secretion (Jacobs, Snyder, Wilbur, Utiger, and Daughaday, 1971; Bowers, Friesen, Hwang, Guyda, and Folkers, 1971). On the other
ROLE IN MAMMOGENESIS AND LACTOGENESIS
177
hand, increasing evidence from in vitro pituitary incubation systems and from studies of human suckling suggests the existence of a hypothalamic factor or factors other than TRH but capable of augmenting PRL secretion. TRH can account for only a portion of the in vitro PRL secretion stimulable by TRH (Machlin et al., 1974), and the reflex release of PRL provoked by suckling is not accompanied by increases in serum TSH concentrations (Gautvik, Tashjian, Kourides, Weintraub, Graeber, Maloof, Suzuki, and Zuckerman, 1974). Circumstantial evidence in both rodents (Lu and Meites, 1973) and man (MacIndoe and Turkington, 1973) supports the hypothesis that neurotransmitter control of such a putative PRH might well be serotonergic rather than dopaminergic, noradrenergic, or cholinergic. The notion that PRL may well be under dual hypothalamic control, despite the quantitative dominance of an inhibitory system, is attractive because of the lack of a known target organ signal capable of completing a closed-loop negative-feedback servo-mechanism. A similar situation may obtain as regards growth hormone, since abundant data demonstrate the dominant in vivo facilitatory effect of the hypothalamus on growth hormone secretion, yet this same hypothalamic tissue also contains substantial quantities of somatostatin. A large number of hormonal and neuropharmacologic influences are capable of modifying hypothalamic control of prolactin secretion. Estrogens cause hyperplasia and probably hypertrophy of the lactotrope cells, and stimulate prolactin synthesis and secretion. This effect is dependent on both the dose and the duration of exposure of estrogens. The modest 2- to 4-fold rises in circulating estradiol-17B during pregnancy in the rhesus monkey are accompanied only by very slight increases in PRL levels (Friesen, Belanger, Guyda, and Hwang, 1972). The mammoth several hundred-fold increase in estradiol during human pregnancy is associated with a 10- to 20fold increase in circulating prolactin (Hwang, Guyda, and Friesen, 1971; Jacobs, Mariz, and Daughaday, 1972). On the other hand, the oral administration of 5 mg of diethylstilbestrol for 1 week to normal young adult men did not raise basal serum PRL levels, but PRL responses to TRH thereafter were enhanced (Carlson, Jacobs, and Daughaday, 1973). Intermediate doses and/or durations of exposure result in intermediate results as regards PRL levels (Frantz, Kleinberg, and Noel, 1972; Yen" Ehara, and Siler, 1974). Since only a minority of women on long-term oral contraceptives have hyperprolactinemia (Daughaday and Jacobs, 1972), however, it seems likely that considerable inter-individual variability exists in response to estrogens. The influence of estr.ogens may be exerted both in the hypothalamus and directly on the pituitary as well. The precise interrelations between the effects of estrogen and of catecholamines on PRL
178
L.S. JACOBS
secretion have not been delineated; nor is it ~ertain whether the effects of estrogen are direct or mediated via neurotransmitter pathways. Similarly, the pathways and mechanisms whereby thyroid hormone deficiency and excess affect PRL secretion are not clearly identified. Hypothyroidism is associated with augmented, and hyperthyroidism with suppressed, PRL secretion (Snyder, Jacobs, Utiger, and Daughaday, 1973). Frequently these alterations are minor in degree, so that basal levels may be normal, although significant departures of group means from those of euthyroid subjects are identified, especially in children. Stimulation with TRH reveals marked augmentation of PRL responses in hypothyroidism and suppression of responses in hyperthyroidism. Since the qualitative nature of these changes is similar to those which occur in the secretion of thyrotropin, since there is general agreement that the direct feedback effects of thyroid hormones on the pituitary are quantatively dominant in TSH feedback control, and since hyperprolactinemia unmasked by TRH occurs in both thyroidal hypothyroidism (Snyder et al., 1973) and in hypothalamic hypothyroidism (Foley et al., 1972), it appears that the common denominator is thyroid hormone deficiency per se. It further seems likely that adenohypophysial lac tot rope cells have thyroid hormone receptors similar to those of the thyrotropes. A variety of drugs have been noted to affect PRL secretion. As noted above, oral contraceptives may cause frank hyperprolactinemia in a minority of women while regular menses continue. Most often, clinically overt galactorrhea does not occur under these circumstances, probably due to the independent action of estrogens to suppress lactational responses to prolactin. The most common drugrelated abnormality of PRL secretion is that which occurs after cessation of oral contraceptive therapy. The relatively common post-pill amenorrhea, in which continued gonadotropin suppression continues inappropriately after cessation of therapy, is on occasion accompanied by hyperprolactinemia, with or without galactorrhea. These clinical manifestations may represent a functional hypothalamic disturbance, or they may herald the presence of a prolactin-secreting pituitary adenoma brought to the threshold of clinical detection by the challenge of oral contraceptive therapy. Other drugs which may cause hyperprolactinemia or galactorrhea do so either by depleting the hypothalamus of catecholamines or by interfering with aminergic transmission in the hypothalamus. Reserpine is the prototype of the former action, and alphamethyldopa, phenothiazines, and butyrophenones interfere with receptor recognition of dopamine. As with oral contraceptives, the overt clinical manifestations--oligomenorrhea or amenorrhea coupled with galactorrhea--probably represent only those individuals with the most marked chemical abnormalities, the tip of the clinical iceberg. With phenothiazines, for example, the overwhelming majority of people taking
ROLE IN MAMMOGENESIS AND LACTOGENESIS
179
significantly more than 50 mg/day equivalent of chlorpromazine will be found to be hyperprolactinemic, yet only a minority will have irregular or absent menses, or galactorrhea. These drugs probably act by limiting the secretion of PIF or by suppressing its action. III.
REGULATION OF MAMMARY GROWTH AND DEVELOPMENT
Our current understanding of the normal sequence of maturational and gestational growth and development of the mammary gland of necessity derives primarily from studies carried out in small laboratory rodents and in dairy animals. Not all the conclusions suggested by these data have been validated in primates. There is general agreement that maturational growth of the mammary gland is probably directly related to prevailing secretory rates of estrogen in vivo. In vitro, the two hormones required for growth and cell division in preparation for prolactin induced milk secretion are insulin and hydrocortisone. In some strains of mice, growth hormone is potently mammotropic (Nandi and Bern, 1961), but human and monkey GH appear to be unique among growth hormones in having lactogenic and mammotropic properties not accountable on the basis of contamination by prolactin. Bovine and ovine GH's are not lactogenic. In organ culture of mouse mammary gland, aldosterone appears to be capable of stimulating ductal differentiation and branching in glands pretreated with insulin and prolactin (Ceriani, 1970), although critical in vivo testing of the need for aldosterone in normal mammogenesis has not been done. No detailed studies of hormonal changes during puberty in experimental animals appear to have been carried out at present. In humans, though in vitro organ culture data are lacking, there is a good deal of information on levels of prolactin and of other potentially relevant hormones at various ages and stages of pubertal and gestational breast development. In addition, a number of hormonally defined disease processes may shed light on the questions at hand. Estrogens stimulate mammary gland hyperplasia in man. Like other estrogen effects, this response is dependent on both the estrogen effects; this response is dependent on both the estrogen level and the duration of tissue exposure to it. Alveolar proliferation, which occurs post-pubertally and is intensified during pregnancy, is highly dependent upon the adequacy of progesterone secretion. It seems reasonable to infer from the organ culture data that normality of cortisol and insulin secretion may be prerequisites for normal mammary growth and development. No systematic clinical data collection on breast development in pubertal girls with Addison's disease has been done, to my knowledge, and juvenile diabetics are treated of course with insulin. Hence, we may have to be satisfied
180
L.S. JACOBS
with guesses and inferences in this matter. It further seems quite likely that euthyroidism may be yet another prerequisite for fully normal breast development. The role played by prolactin in mammogenesis in humans remains unclear. Circulating levels of prolactin are only slightly higher in adult women of reproductive age than in men, and measurements in prepubertal children have generally given results comparable to those in men (Foley et al., 1972; Guyda and Friesen, 1973). In cross-sectional studies of puberty, the observed rise in PRL levels seemed to be gradual (Guyda and Friesen, 1973). Similarly, the rise in estradiol levels appears to be gradual during puberty, with achievement of adult female levels only late in the pubertal process (Ehara, Yen, and Siler, 1975), whereas breast budding and major breast growth in girls are relatively early pubertal events. These data do not support the concept that major changes in estrogen or prolactin levels are responsible for human mammogenesis. Rather, it seems likely that subtle changes in hormonal secretion, coupled with alterations in mammary responsiveness to the hormonal milieu, may be responsible for human pubertal breast growth. It seems reasonable to assume that certain minimal levels of prolactin may be required, in a permissive way, for breast development to occur The occurrence of hypopituitarism or hypogonadism prepubertally is associated with deficient mammary development in the absence of replacement therapy. In hypopituitary girls, treatment with glucocorticoids, thyroid hormone, and growth hormone does not result in notable mammary development. This requires treatment with estrogen, estrogen plus progesterone, or gonadotropins. Prolactin secretion is frequently normal in hypopituitary subjects. In girls with Turner's syndrome, breast development requires sex steroid therapy. Women with isolated growth hormone deficiency develop normal breasts and are fully capable of normal postpartum lactational performance (Tyson, Hwang, Guyda, and Friesen, 1972; Rimoin, Holzman, Merimee, Rabinowitz, Barnes, Tyson, and McKusick, 1968). These considerations strongly suggest that in humans, although eucorticism and euthyroidism may be required for the full expression of mammary growth, maturation, and lactational function, they exert a permissive role only. Growth hormone appears not to play an important role. Since most early breast growth is accounted for by ductal proliferation without much alveolar budding in normal girls whose aldosterone secretion is adequate, one may also question the relevance of the mouse mammary organ culture data cited above to the normal human situation. Although systematic observations in man on the role of androgens in mammary development have not been made, it is clear from
ROLE IN MAMMOGENESIS AND LACTOGENESIS
181
clinical observations in women with virilizing syndromes and in men with alcoholism and gynecomastia that androgens act in opposition to estrogens as regards breast development. No comparison of PRL levels or estradiol levels between buxom and flat-chested women have been made. Table 1 lists the hormones which may be involved in mammary growth and development. The growth and development of the breast during pregnancy is accounted for by the combined effects of hypersecretion of estrogens, progesterone, prolactin, and chorionic somatomammotropin. Considerable alveolar hypertrophy occurs during pregnancy, and if it were not for the inhibitory effect of very high circulating estrogens on the breast response to mammotropic hormones, undoubtedly copious lactation would be a regular feature of the third trimester of pregnancy. As it is, the hormonal bombardment of the breast in late pregnancy is quite remarkable; estradiol and progesterone levels are several hundred-fold higher in late pregnancy than in the follicular phase of the menstrual cycle. Chorionic somatomammotropin levels may rise as high as 10-12 ~g/ml, several orders of magnitude higher than the usual circulating levels of other peptide hormones of comparable molecular weight. Pituitary prolactin levels, under the influence of estrogens, average about 200 ng/ml at term, or roughly twenty times higher than nonpregnant levels. IV.
REGULATION OF MILK PRODUCTION
I will not review here the extensive literature on regulation of lactation in dairy animals, but will rather focus on what is known regarding this subject in women. Many of the same hormonal influences important in the regulation of mammary growth and development play prominent roles in the control of lactation. Failure of lactation in otherwise endocrinologically normal women has been described in association with isolated deficiency of prolactin (Turkington, 1972); these data were obtained by PRL bioassay measurement following administration of phenothiazines, and have yet to be confirmed. From the in vitro organ culture experience, one may suspect that normal insulin secretion, eucorticism, and euthyroidism are important permissive prerequisites for normal lactation. The secretory apparatus must have reached some minimal threshold stage of development under the influence of estrogen and progesterone. Induction of milk protein synthesis is dependent on the mammotropic hormones prolactin or chorionic somatomammotropin. The secretion of milk following all this hormonal preparation further depends on withdrawal of estrogen and progesterone, as well as on intact mechanisms for oxytocin release and myoepithelial basket cell contraction. Tactile stimulation of the breast and nipple results in reflex release of PRL (Kolodny, Jacobs, and Daughaday, 1972; Noel, Suh, and Frantz, 1974) and of oxytocin as well. In endocrinologically normal but nonpregnant women, lactogenesis can be induced by administration of estrogen and a regular program of tac-
L.S. JACOBS
182
tile manipulation of the breast and nipple, followed by abrupt cessation of the estrogen treatment and a waiting period of a few days (Tyson, Khojandi, Ruth, and Andreasson, 1975). Similarly, there is a normal delay of several days following delivery before milk secretion becomes well established postpartum. The important factors responsible for this delay include time required to escape from the inhibition previously exercised by high levels of estrogen and progesterone. With delivery of the placenta, the source of all the placental lactogen (chorionic somatomammotropin) and the great bulk of the progesterone and estrogen produced during late pregnancy is removed. Circulating levels of these hormones fall in accord with their respective half-lives in the serum, but the secretion of prolactin does not change nearly as rapidly. Even in nonbreast-feeding mothers experiencing no suckling stimulus, it may take 2 to 3 weeks for basal prolactin levels to return to nonpregnant levels. This indicates substantial temporal persistence of the estrogen stimulation after its removal, and is consistent with the duration of other biological actions of estrogens. In breast-feeding mothers, the decline of PRL levels is substantially slowed, and nonpregnant fasting levels may not be reached for several weeks. Table 2 summarizes the factors involved in the initiation of postpartum lactation. As has been shown, suckling episodes result in dramatic rises in serum PRL levels during the early puerperium (Tyson et al., 1972). Later during lactation, the serum PRL either increases much less sharply with suckling, or in some cases may not rise measurably at all; basal PRL levels are normal. Although there is some variation in serum PRL from moment to moment, especially in hyperprolactinemic states (Jacobs and Daughaday, 1973), there is reason to believe that randomly obtained levels may nonetheless reflect integrated 24-hour levels reasonably Well (Malarkey, 1975). Thus normal lactation seems to proceed quite well in the absence of accelerated PRL secretion. At the same time, most breast-feeding women remain amenorrheic or oligomenorrheic and rather severely estrogen-deficient. Nonpuerperal galactorrhea is also frequently associated with hypoestrogenism and menstrual irregularities or amenorrhea. Thus, in contrast to breast development, which requires estrogen, lactation seems to require suppression or removal of estrogen. Further, it would appear that Table 2.
FACTORS INVOLVED IN TRE INITIATION OF POSTPARTUM LACTATION Postpartum Time Lag Rapidly Falling Estrogens Rapidly Falling Progesterone Tactile Stimulation of the Breast and Nipple
ROLE IN MAMMOGENESIS AND LACTOGENESIS Table 3.
183
ASSESSMENT OF PROLACTIN EFFECTS ON THE BREAST
Single Determinations VB. Secretory Rates Subtle alterations Maintained over Months or Years Synergistic Effects of Multiple Hormones Variations in Mammary Response to the Hormonal Milieu Systemic VB. Mammary Hormone Concentrations more PRL is required for the initiation of lactation than for its maintenance. Table 3 indicates some of the considerations which need to be kept in mind as one tries to relate measured PRL levels to breast function. The possibility does of course exist that the measured serum PRL level is not an accurate reflection of the hormonal milieu to which the breast is exposed. In order to assess this possibility, and because the presence of immunoreactive PRL in milk had been reported in experimental animals (McMurtry and Malvern, 1974a, 1974b), we have measured the PRL in simultaneously obtained milk and serum specimens in a series of women with nonpuerperal galactorrhea. Figure 1 shows the systematic gradient which exists between these two fluids, the prolactin concentration in milk being significantly higher than that found in serum. Thus the breast does to a certain extent act as a concentrating mechanism. Whether the PRL in milk which is immunologically intact is a more valid index of the hormonal exposure of milk-secreting cells than that in serum remains to be shown, however. Although the mechanism by which apparently intact prolactin finds its say into milk is not known, it seems clear that the hormone must traverse the capillary side of the alveolar cell plasma membrane, the cell interior, a~d the luminal or ductular surface plasma membrane as well. Cell penetrance by polypeptide hormones such as prolactin has in the past been thought not to occur in relation to hormone action; however, recently intracellular polypeptide hormone receptors have been tentatively identified (Posner and Bergeron, 1976) and such a mechanism may be involved in the transfer of prolactin from serum to milk. A similar gradient in PRL seems to exist in postpartum lactating women, so this mechanism is not restricted to situations of abnormal pathophysiology. It is a reasonably frequent occurrence ot find normal PRL levels in the serum of women with nonpuerperal galactorrhea of diverse etiologies. This is especially common in women in whom regular menses persist despite galactorrhea, and no apparent cause for the lactation can be found. Several possible explanations exist for continued galactorrhea despite normal PRL levels. One may postulate that during the early phases of the disorder, higher levels of PRL may have prevailed while lactation was being established.
L.S. JACOBS
184
230 220 140 120
E ...... 0> c: 100 Z I-
u
«
80
-I
0
~
a..
:x: -I
~
60 40 20
o Figure 1.
20
40
60
80
100
120
SERUM PROLACTIN (ng/ml)
Correlation between serum PRL levels and simultaneously obtained milk PRL levels in women with nonpuerperal galactorrhea. The bulk of the data points fall above the diagonal line of identity, indicating a systematic gradient from milk to serum. Whether enhanced entry into, or diminished clearance from milk is the prime factor responsible for the higher milk PRL levels is not known.
ROLE IN MAMMOGENESIS AND LACTOGENESIS Table 4.
185
FACTORS CAPABLE OF INFLUENCING MAMMARY RESPONSIVITY TO PROLACTIN I. II.
III.
PRE-RECEPTOR EVENTS Prior Hormonal Milieu THE PROLACTIN RECEPTOR Affinity Capability POST-RECEPTOR EVENTS Coupling of Hormone Binding to Adeny1ate Cyclase Protein Kinase Phosphorylation of Intracellular Proteins Km, Vmax , of Enzymes Involved in Milk Fat and/or Carbohydrate Synthesis
Alternatively, one may explain such a situation by the suggestion that milk PRL levels may be higher than those in serum, as shown in Figure 1. Finally, an altered, heightened tissue responsivity to PRL could account for such findings. One way in which heightened tissue sensitivity may be explained is by an increase in either the number or the affinity of hormone receptors. Figure 2 demonstrates that just such an increase in PRL affinity does in fact occur in rabbit mammary glands during postpartum lactation. Individual Scatchard plots of PRL to representative early (first postpartum week) and last (just prior to weaning) mammary gland membrane preparations are shown. Although the mechanism responsible for this 2.5-fo1d increase in PRL receptor affinity is not known, such as increase could help account for continued lactation in the face of falling serum PRL levels. Table 4 indicates that changes in hormone receptor affinity represent only one of several possible mechanisms which might underlie enhanced tissue responsivity. On occasion, despite very high serum PRL levels, sometimes in the thousands of nanograms per m1, no mammary manifestations attributable to hyperpro1actinemia may occur. This type of dissociation between clinical manifestations and hormone overproduction is most often seen in patients harboring pituitary adenomas. The absence of gynecomastia in such men, and of galactorrhea in such women, is
L.S. JACOBS
186
0.4
0.3
B/F 0.2
0.1 Early /Pre p #1
I
o
o
1 2 3 456 ng oPRL BOUND I mg PROTEIN I
I
I
I
I
I
52.2 104.4 157 209 261 313 fmoles oPRL BOUND/mg PROTEIN
Figure 2.
Scat chard plots of the interaction of single representative early (first postpartum week) and late (immediately prior to weaning) mammary receptor preps with iodinated ovine prolactin. The binding capacity of the two preparations is similar (60 and 76 fmoles/mg protein) but the affinity of the late prep is 3.8-fold greater than that of the early prep (Ka,l x 1010M-1VS. 2.6 x 109 M- I) .
ROLE IN MAMMOGENESIS AND LACTOGENESIS
187
most satisfactorily explained by postulating that the frequently present hypogonadism results in hypoestrogenism and breasts which are less sensitive to the actions of PRL. Most men with PRL-secreting pituitary adenomas who do have galactorrhea exhibit serum PRL levels which are very markedly elevated, often in excess of 1000 ng/ml. Although many women with such tumors and galactorrhea have PRL levels which are only modestly elevated, suggesting that the clinical expression of hyperprolactinemia at the breast is facilitated by estrogp-ns, nonetheless men with such tumors occasionally have galactorrhea with PRL levels less than 100 ng/ml. Uniformly these men are found to be hypogonadal. Thus it would seem, as in cirrhotics with gynecomastia, that breast responses to sex steroids and to PRL appear to depend on the estrogen/androgen ratio more than the absolute level of either. REFERENCES Ben-Jonathan, N., Oliver, C, and Mical, R.S. (1976). Dopamine secretion into hypophysial portal blood during the estrus cycle and pregnancy in the rat. Program, 58th Meeting, Endocrine Society, Abstract 269, Endocrinology 98, Supplement, p. 191. Birge, C.A., Jacobs, L.S., Hammer, C.T., and Daughaday, W.H. (1970). Catecholamine inhibition of prolactin secretion by isolated rat adenohypophyses. Endocrinology 86, 120-130. Bowers, C.Y., Friesen, H.G., Hwang, P., Guyda, H.J., and Folkers, K. (1971). Prolactin and thyrotropin release in man by synthetic pyroglutamyl-histidyl-prolinamide. Biochem. Biophys. Res. Commun. 45, 1033-1041. Carlson, H.E., Jacobs, L.S., and Daughaday, W.H. (1973). Growth hormone, thyrotropin and prolactin responses to thyrotropinreleasing hormone following diethylstilbestrol pretreatment. J. Clin. Endocrinol. Metab. 37, 488-490. Ceriani, R.L. (1970) Fetal mammary gland differentiation in vitro in response to hormones. I. Morphological findings. Developm. Biol. 21, 506-529. Daughaday, W.H. and Jacobs, L.S. (1972). Normal and pathologic secretion of prolactin in man. In: Endocrinology, Proceedings IV International Congress of Endocrinology (Scow, R.O., ed.), pp. 622-628, Excerpta Medica International Congress Series No. 273. del Pozo, E., Varga, L., Wyss, H., Tolis, G., Friesen, H., Wenner, R., Vetter, L., and Uettwiler, A. (1974). Clinical and hormonal response to bromocriptin (CB-154) in the galactorrhea syndromes. J. Clin. Endocrinol. Metab. 39, 18-36. Ehara, Y., Yen, S.S.C., and Siler, T.M. (1975). Serum prolactin levels during puberty. Amer. J. Obstet. Gynec. 121, 995-997.
L.S. JACOBS
188
Foley, T.P. Jr., Jacobs, L.S., Hoffman, W., Daughaday, W.H., and Blizzard, R.M. (1972). Human prolactin and thyrotropin concentrations in the serums of normal and hypopituitary children before and after the administration of synthetic thyrotropinreleasing hormone. J. Clin. Invest. 51, 2143-2150. Frantz, A.G., Kleinberg, D.L., and Noel, G.L. (1972). Studies on prolactin in man. Rec. Prog. Horm. Res. 28, 527-590. Friesen, H.G., Belanger, C., Guyda, H.J., and Hwang, P. (1972). The synthesis and secretion of placental lactogen and pituitary prolactin. In: Lactogenic HOPmones (Wolstenholme, G.F. W. and Knight, J., eds.), pp. 83-103, Churchill-Livingstone, London. Fuxe, K. (1964). Cellular localization of monoamines in the median eminence and the infundibular stem of some mammals. Z. Zellforsch Mikrosk. Anat. 61, 710-724. Gautvik, K.M., Tashjian, A.H. Jr., Kourides, I.A., Weintraub, B.D., Graeber, C.T., Maloof, F., Suzuki, K., and Zuckerman, J.E. (1974). Thyrotropin-releasing hormone is not the sole physiologic mediator of prolactin release during suckling. New Engl. J. Med. 290, 1162-1165. Hwang, P., Guyda, H., and Friesen, H. (1971). A radioimmunossay for human prolactin. Proc. Natl. Acad. Sci. 68, 1902-1906. Jacobs, L.S. and Daughaday, W.H. (1973). Pathophysiology and control of prol~c.tin sec.retion in patients with pituitary and hypothalamic disease. In: Human Prolactin (Pasteels, J.L. and Robyn, C., eds.), pp. 189, International Congress Series No. 308, Excerpta Medica, Amsterdam. Jacobs, L.S., Mariz, I.K., and Daughaday, W.H. (1972). A mixed heterologous radioimmunoassay for human prolactin. J. Clin. Endocrinol. Metab. 34, 484-490. Jacobs, L.S., Snyder, P.J., Wilber, J.F., Utiger, R.D., and Daughaday, W.H. (1971). Increased serum prolactin after administration of synthetic thyrotropin releasing hormone (TRH) in man. J. Clin. Endocrinol. Metab. 33, 996-999. Kolodny, R.C., Jacobs, L.S., and Daughaday, W.A. (1972). Mammary stimulation causes prolactin secretion in non-lactating women. Nature 238, 284-286. Larson, B.L. and Smith, V.R., eds. (1974). Lactation/A Comprehensive Treatise, 3 volumes, Academic Press. Volume 1: The Mammary Gland/Development and Maintenance, Volume 2: Biosynthesis and Secretion of Milk/Diseases, and Volume 3: Nutrition and
Biochemistry of Milk/Maintenance of Lactation.
Lu, K.H. and Meites, J. (1973). Effects of serotonin precursors and melatonin on serum prolactin release in rats. Endocrinology 93, 152-155. MachI in , L.J., Jacobs, L.S., Cirulis, N., Kimes, R., and Miller, R. (1974). An assay for growth hormone and prolactin-releasing activities using a bovine pituitary cell culture system. Endocrinology 95, 1350-1358.
ROLE IN MAMMOGENESIS AND LACTOGENESIS
189
Maclndoe, J.H. and Turkington, R.W. (1973). Stimulation of human prolactin secretion by intravenous infusion of L-tryptophan. J. Clin. Invest. 52, 1972-1978. Macleod, R.M. and Lelli~eyer, J.E. (1974). Studies on the mechanisms of the dopamine-mediated inhibition of prolactin secretion. Endocrinology 94, 1077-1085. Malarkey, W.B. (1975). Nonpuerperal lactation and normal prolactin regulation. J. Clin. Endocrinol. Metab. 40, 198-204. McMurtry, J.P. and Malven, P.V. (1974a). Radioimmunoassay of endogenous and exogenous prolactin in milk of rats. J. Endocrinol. 61, 211-217. McMurtry, J.P. and Malven, P.V. (1974b). Experimental alterations of prolactin levels in goat milk and blood plasma. Endocrinology 95, 559-564. Meites, J. and Nicoll, C.S. (1966). Adenohypophysis: Prolactin. Ann. Rev. Physiol. 28, 57-88. Nandi, S. and Bern, H.A. (1961). The hormone responsible for lactogenesis in BALB-cCrgl mice. Gen. Compo Endocrinol. 1, 195210. Nicoll, C.S., Fiorindo, R.P., McKennee, C.R., and Parsons, J.A. (1970). Assay of hypothalamic factors which regulate prolactin secretion. In: Hypophysiotropic Hormones of the Hypothalamus: Assay and Chemistry (Meites, J., ed.), pp. 115-144, Williams and Wilkins, Baltimore. Noel, G.L., Suh, H.K., and Frantz, A.G. (1974). Prolactin release during nursing and breast stimulation in postpartum and nonpostpartum subjects. J. Clin. Endocrinol. Metab. 38, 413-423. Posner, B.I. and Bergeron, J.J.M. (1976). Intracellular polypeptide hormone receptors. Program, 58th Meeting, Endocrine Society, Abstract 218, Endocrinology 98, Supplement, pp. 165. Rimoin, D.L., Holzman, G.B., Merimee, T.J., Rabinowitz, D., Barnes, A.C., Tyson, J.E., and McKusick, V.A. (1968). Lactation in the absence of human growth hormone. J. Clin. Endocrinol. Metab. 28, 1183-1188. Shaar, C.J. and Clemens, J.A. (1974). The role of catecholamines in the release of anterior pituitary prolactin in vitro. Endocrinology 95, 1201-1212. Snyder, P.J., Jacobs, L.S., Utiger, R.D., and Daughaday, W.H. (1973). Thyroid hormone inhibition of the prolactin response to thyrotropin-releasing hormone. J. Clin. Invest. 52, 2324-2329. Snyder, P.J., Jacobs, L.S., Rabello, M.M., Sterling, F.H., Shore, R.N., Utiger, R.D., and Daughaday, W.H. (1974). Diagnostic value of thyrotropin-releasing hormone in pituitary and hypothalamic diseases: Assessment of thyrLtropin and prolactin secretion in 100 patients. Annals Int. Med. 81, 751-757. Turkington, R.W. (1972). Phenothiazine stimulation test for prolactin reserve: The syndrome of isolated prolactin deficiency. J. Clin. Endocrinol. Metab. 34, 247-250.
L.S. JACOBS
190
Tyson, J.E., Hwang, P., Guyda, H., and Friesen, H.G. (1972). Studies of prolactin secretion in human pregnancy. Amer. J. Obstet. Gynecol. 113, 14-20. Tyson, J.E., Khojandi, M., Huth, J., and Andreasson, B. (1975). The influence of prolactin secretion on human lactation. J. Clin. Endocrinol. Metab. 40, 764-773. Valverde, C., and Chieffo, V. (1971). Prolactin releasing factors in porcine hypothalamic extracts. Program, 53rd Meeting, Endocrine Society, Abstract 83, Endocrinology 88, Supplement, p. A84. Woolf, P.D., Jacobs, L.S., Donofrio, R., Burday, S.Z., and Schalch, D.S. (1974). Secondary hypopituitarism: Evidence for continuing regulation of hormone release. J. Clin. Endocrinol. Metab. 38, 71-76. Yen, S.S.C., Ehara, Y., and Siler, T.M. (1974). Augmentation of prolactin secretion by estrogen in hypogonadal women. J. Clin. Invest. 53, 652-655. DISCUSSION AFTER DR. JACOBS' PAPER Dr. Barna1.Je U I wUlulel- if you know af any measurement R of prolact in levels
during the menstrual cycle. Dr. Jacobs
Yes, several groups of investigators have measured prolactin levels in a very large number of normally cycling women. Although one group of European investigators claim that there is a significant increase in prolactin during the luteal phase, I think by and large most investigators have found no significant differences, so I would say that there is no significant variation in prolactin during the menstrual cycle. Dr. Barna1.Je U
Except for the possibility that nocturnal levels might be higher.
Dr. Jacobs
Yes, that remains a possibility.
Dr. Klein
Dr. Yen published data which showed that the nocturnal rise in prolactin at approximately mid-cycle is about twice as high as during either the follicular or luteal phase, but that's on only one patient (Ehara, Siler, Vandenberg, Sinha, and Yen, 1973). Has anyone investigated the biological activity of the prolactin found in some of the patients with tumors who don't lactate? Could this be immunoassayable prolactin that is not biologically active?
ROLE IN MAMMOGENESIS AND LACTOGENESIS
191
Dr. Jacobs
Some of the best data on this question were shown by Dr. Frantz in terms of the good general correlation between bioassay and radioimmunoassay results. In addition, Dr. Friesen has shown excellent correlation between radioimmunoassay and radioreceptor assay results in a large number of specimens spanning both the physiologic and pharmacologic ranges (Friesen, 1973). These data suggest strongly that the failure to observe lactation in many women with prolactinproducing pituitary adenomas is the result of variable breast responses rather than biologically inactive hormone.
Dr. Sage
I'm not familiar with the clinical literature, but the La Leche league reports on a number of cases of women who have never had children, but who adopt children and are capable of producing milk. Have any of these been followed clinically and the prolactin levels assayed?
Dr. Jacobs
Not to my knowledge. It is well documented, however, that postmenopausal grandmothers have successfully suckled their grandchildren. This used to be common in certain African tribal communities in which the men hunted and the reproductive age women did all the agricultural field work. The critical factor here, and in the La Leche reports to which you refer, is probably tactile stimulation of the breast and especially the nipple. Such tactile stimulation can, reflexly, cause release of prolactin. The reflex appears to be estrogen-sensitive. I might add a fascinating bit of what one might call "mammaryology" in the marsupial. Kangaroos not infrequently give birth while still breastfeeding, and it is not unusual, I understand, for the rather advanced joey at age one or more to be breastfeeding at the same time that there is a newborn in the pouch breastfeeding. It has been shown that the composition of the milk produced by the two teats is markedly different. There are differences in protein electrophoretic patterns and differences in fat composition. Since they corne from the same mother, one would presume that the hormonal milieu is the same. Hence, the observed differences in composition may relate to differences in quantity, quality, or frequency of the tactile stimulation produced by the neonate and that of the joey. DISCUSSION REFERENCES Ehara, Y., Siler, T., Vandenberg, G., Sinha, Y.M., and Yen, S.S.C. (1973). Circulating prolactin levels during the menstrual cycle: Episodic release and diurnal variation. Amer. J. Obstet. Gynec. 117, 962-970. Friesen, H.G. (1973). In Human Prolactin (Pasteels, J.L. and Robyn, C., eds.), Ies #308, Excerpta Medica, Amsterdam.
Laurence S. Jacobs Washington University School of Medicine Department of Medicine-Metabolism 660 S. Euclid St. Louis, Missouri 63110 OUTLINE I. II.
Introduction Control of Prolactin Secretion Hypothalamic influences Estrogen Thyroid function Drugs
III.
Regulation of Mammary Growth and Development Normal hormonal requirements Insights gained from disease states Effects of pregnancy
IV.
Regulation of Milk Production Determinants of postpartum lactation Relationship between serum PRL levels and lactation Milk PRL levels PRL receptor affinity Interactions with steroids I.
INTRODUCTION
Although a great deal of new information about the physiologic and pathologic regulation of prolactin secretion in man has become available in recent years, precise delineation of the role of prolactin in human mammogenesis and lactogenesis has been difficult. This should not be surprising in light of the fact that its role in these processes is imperfectly understood in sub-primate mammals.
173 H.-D. Dellmann et al. (eds.), Comparative Endocrinology of Prolactin © Plenum Press, New York 1977
L.S, JACOBS
174
Though detailed information has been accumulated 011 hormonal requirements for the maintenance and function of mammary explants in short-term tissue culture in vitro, the relevance of these observations to in vivo lactational physiology remains unclear. A substantial body of literature dealing with lactation and its vascular cytologic, biochemical, and physiologic aspects has been developed during the past 30 years, but many aspects of breast function remain poorly understood (Larson and Smith, 1974). At the outset, it should be noted that many reasons for the relative paucity of our understanding of breast physiology are apparent. One of the most important of these is the large number of hormonal signals to which the breast appears to be responsive. Estrogen, progesterone, placental lactogen, and pituitary prolactin play important roles in determining breast size and in modulating breast function. Additionally, there is substantial albeit less compelling existence that lactational performance is in part dependent on normal thyroid and adrenal function; softer evidence further indicates the likelihood that the mammary gland requires insulin for normal functioning. The influence of these latter factors on normal breast development remains essentially unknown. We cannot extrapolate with very much certainty from the rodent models which have served as the basis for much of the in vitro information to the human lactational situation. Because of the dominant inhibitory hypothalamic control over prolactin secretion, lesions of the stalk or hypothalamus in man which do not also totally destroy the lactotrope cell population may cause marked increases in prolactin secretion (Foley, Jacobs,
Table 1.
HORMONAL FACTORS IN HUMAN MAMMARY GROWTH AND DEVELOPMENT I.
II.
III.
IV.
Hormones of Established Prime Importance Estrogens Progesterone Prolactin Placental Lactogen (chorionic somatomammotropin) Hormones Probably Required in a Permissive Way Thyroxine/Triiodothyronine Insulin Cortisol Hormones Implicated by Animal Studies but Probably not Relevant to Human Mammogenesis Growth Hormone Aldosterone Hormones Inhibiting Breast Development Androgens
ROLE IN MAMMOGENESIS AND LACTOGENESIS
175
Hoffman, Daughaday, and Blizzard, 1972; Snyder, Jacobs, Rabello, Sterling, Shore, Utiger, and Daughaday, 1974). One consequence of this fact is that it is hazardous to assume ablation of PRL 1 secretion following lesions of the hypothalamic-pituitary axis. Rather, empirical validation of PRL deficiency, based on radioimmunoassay measurement of circulating hormone levels, is required. Further, measurement of basal levels is insufficient. It should also be demonstrated that administration of a potent stimulus to PRL secretion such as TRH, which acts directly on the adenohypophysial lactotrope cell (MachI in , Jacobs, Cirulis, Kimes, and Miller, 1974), fails to elicit an increase in PRL secretion before one may safely conclude that hypolactotropism has been achieved. These considerations loom large when one considers the type of experimental evidence which would be required for the validation of an experimental model of PRL deficiency. An even greater paucity of useful information exists regarding the role of prolactin in the initiation and control of mammogenesis in man. The difficulties in advancing our understanding of this problem are related primarily to the lack of a suitable experimental model. Two features contribute: first, the scarcity and expense of a suitable primate species for study; second, the technical difficulties in producing long-term isolated deficiency of prolactin without concomitantly affecting the secretion of other pituitary hormones and their target organ secretions. Nonetheless, experimental evidence supporting a major role for prolactin in animal mammogenesis is compelling (Meites and Nicoll, 1966). In order to study the role of the hormone in mammogenesis, one should choose for study an animal in which mammary growth and development is readily identified as a normal maturational event. It would be preferable to establish a state of PRL deficiency in early postnatal life, and it would be essential of course that the PRL deficiency be an isolated one. These considerations make it difficult to draw conclusions regarding the physiologic role of prolactin in an animal, like some strains of mice, in whom growth hormone is well documented to have a potent mammotropic effect. This brief sketch of the formidable technical barriers which have thus far impeded more rapid progress in our understanding of breast maturation and function should serve as a background for the ensuing discussion of the control of prolactin secretion. We shall return later to the question of experimental approaches to the un-
lAbbreviations used in this manuscript are: PRL: prolactin; TRH: thyrotropin releasing hormone; PIF: prolactin inhibiting factor; TSH: thyrotropin; PRH: prolactin releasing hormone; GH: growth hormone.
L.S. JACOBS
176 derstanding of human breast maturation and function.
After even the briefest consideration of some of these issues it is not hard to understand why we have yet to see a definitive experimental study of the role of prolactin in in vivo mammogenesis. Total ablation of adenohypophysial endocrine functions, followed by fractional replacement therapy with thyroid hormone, glucocorticoids, sex steroids, and growth hormone, might shed some light on this problem, but would be a tedious and difficult longterm experiment. II.
CONTROL OF PROLACTIN SECRETION
Unlike the remainder of the anterior pituitary hormones in man, prolactin is regulated in vivo by a dominant inhibitory influence of the hypothalamus. Although it is clear from a large body of experimental evidence that dopaminergic mechanisms are involved in the inhibition of PRL secretion, the major locus at which dopamine may act in vivo is not clear. Considerable data in animal experiments as well as in man point to an effect of dopamine to suppress PRL secretion directly from the adenohypophysis (Birge, Jacobs, Hammer, and Daughaday, 1970; Macleod and Lehmeyer, 1974; Woolf, Jacobs, Donofrio, Burday, and Schalch, 1974; Shaar and Clemens, 1974). However, the hypothalamus is very rich in dopamine (Fuxe, 1974) and this compound is found in hypophysial portal vessel blood (Ben-Jonathan, Oliver, and Mical, 1976). It is conceivable that dopaminergic mechanisms may operate in vivo on both the hypothalamus and the pituitary. In the clinical arena, agents which are dopamine agonists, such as the precursor L-dopa or the ergot derivative 2-a-bromoergocryptine, have been used to suppress prolactin secretion and restore menses and fertility in hyperprolactinemic-amenorrheic women (del Pozo, Varga, Wyss, Tolis, Friesen, Wenner, Vetter, and Uettwiler, 1974). At present, it is not clear whether a prolactin inhibiting factor (PIF) of peptide nature exists as a hypothalamic entity separate from dopamine or not (Shaar and Clemens, 1974). In addition to the inhibitory effect which is readily unmasked
in vivo by experimental transection of the pituitary stalk, evidence
has accumulated that a quantitatively less important hypothalamic stimulatQry effect on PRL secretion may also exist. Data in support of this concept have come from studies on avian (Nicoll, Fiorindo, McKennee, and Parsons, 1970) and porcine (Valverde and Chieffo, 1971) hypothalamus. The biochemical interpretation most appropriate to these physiologic data renlain unclear, however. It has been established now for half a decade that TRH, a tripeptide of known hypothalamic residence, is a potent stimulator of PRL as well as of TSH secretion (Jacobs, Snyder, Wilbur, Utiger, and Daughaday, 1971; Bowers, Friesen, Hwang, Guyda, and Folkers, 1971). On the other
ROLE IN MAMMOGENESIS AND LACTOGENESIS
177
hand, increasing evidence from in vitro pituitary incubation systems and from studies of human suckling suggests the existence of a hypothalamic factor or factors other than TRH but capable of augmenting PRL secretion. TRH can account for only a portion of the in vitro PRL secretion stimulable by TRH (Machlin et al., 1974), and the reflex release of PRL provoked by suckling is not accompanied by increases in serum TSH concentrations (Gautvik, Tashjian, Kourides, Weintraub, Graeber, Maloof, Suzuki, and Zuckerman, 1974). Circumstantial evidence in both rodents (Lu and Meites, 1973) and man (MacIndoe and Turkington, 1973) supports the hypothesis that neurotransmitter control of such a putative PRH might well be serotonergic rather than dopaminergic, noradrenergic, or cholinergic. The notion that PRL may well be under dual hypothalamic control, despite the quantitative dominance of an inhibitory system, is attractive because of the lack of a known target organ signal capable of completing a closed-loop negative-feedback servo-mechanism. A similar situation may obtain as regards growth hormone, since abundant data demonstrate the dominant in vivo facilitatory effect of the hypothalamus on growth hormone secretion, yet this same hypothalamic tissue also contains substantial quantities of somatostatin. A large number of hormonal and neuropharmacologic influences are capable of modifying hypothalamic control of prolactin secretion. Estrogens cause hyperplasia and probably hypertrophy of the lactotrope cells, and stimulate prolactin synthesis and secretion. This effect is dependent on both the dose and the duration of exposure of estrogens. The modest 2- to 4-fold rises in circulating estradiol-17B during pregnancy in the rhesus monkey are accompanied only by very slight increases in PRL levels (Friesen, Belanger, Guyda, and Hwang, 1972). The mammoth several hundred-fold increase in estradiol during human pregnancy is associated with a 10- to 20fold increase in circulating prolactin (Hwang, Guyda, and Friesen, 1971; Jacobs, Mariz, and Daughaday, 1972). On the other hand, the oral administration of 5 mg of diethylstilbestrol for 1 week to normal young adult men did not raise basal serum PRL levels, but PRL responses to TRH thereafter were enhanced (Carlson, Jacobs, and Daughaday, 1973). Intermediate doses and/or durations of exposure result in intermediate results as regards PRL levels (Frantz, Kleinberg, and Noel, 1972; Yen" Ehara, and Siler, 1974). Since only a minority of women on long-term oral contraceptives have hyperprolactinemia (Daughaday and Jacobs, 1972), however, it seems likely that considerable inter-individual variability exists in response to estrogens. The influence of estr.ogens may be exerted both in the hypothalamus and directly on the pituitary as well. The precise interrelations between the effects of estrogen and of catecholamines on PRL
178
L.S. JACOBS
secretion have not been delineated; nor is it ~ertain whether the effects of estrogen are direct or mediated via neurotransmitter pathways. Similarly, the pathways and mechanisms whereby thyroid hormone deficiency and excess affect PRL secretion are not clearly identified. Hypothyroidism is associated with augmented, and hyperthyroidism with suppressed, PRL secretion (Snyder, Jacobs, Utiger, and Daughaday, 1973). Frequently these alterations are minor in degree, so that basal levels may be normal, although significant departures of group means from those of euthyroid subjects are identified, especially in children. Stimulation with TRH reveals marked augmentation of PRL responses in hypothyroidism and suppression of responses in hyperthyroidism. Since the qualitative nature of these changes is similar to those which occur in the secretion of thyrotropin, since there is general agreement that the direct feedback effects of thyroid hormones on the pituitary are quantatively dominant in TSH feedback control, and since hyperprolactinemia unmasked by TRH occurs in both thyroidal hypothyroidism (Snyder et al., 1973) and in hypothalamic hypothyroidism (Foley et al., 1972), it appears that the common denominator is thyroid hormone deficiency per se. It further seems likely that adenohypophysial lac tot rope cells have thyroid hormone receptors similar to those of the thyrotropes. A variety of drugs have been noted to affect PRL secretion. As noted above, oral contraceptives may cause frank hyperprolactinemia in a minority of women while regular menses continue. Most often, clinically overt galactorrhea does not occur under these circumstances, probably due to the independent action of estrogens to suppress lactational responses to prolactin. The most common drugrelated abnormality of PRL secretion is that which occurs after cessation of oral contraceptive therapy. The relatively common post-pill amenorrhea, in which continued gonadotropin suppression continues inappropriately after cessation of therapy, is on occasion accompanied by hyperprolactinemia, with or without galactorrhea. These clinical manifestations may represent a functional hypothalamic disturbance, or they may herald the presence of a prolactin-secreting pituitary adenoma brought to the threshold of clinical detection by the challenge of oral contraceptive therapy. Other drugs which may cause hyperprolactinemia or galactorrhea do so either by depleting the hypothalamus of catecholamines or by interfering with aminergic transmission in the hypothalamus. Reserpine is the prototype of the former action, and alphamethyldopa, phenothiazines, and butyrophenones interfere with receptor recognition of dopamine. As with oral contraceptives, the overt clinical manifestations--oligomenorrhea or amenorrhea coupled with galactorrhea--probably represent only those individuals with the most marked chemical abnormalities, the tip of the clinical iceberg. With phenothiazines, for example, the overwhelming majority of people taking
ROLE IN MAMMOGENESIS AND LACTOGENESIS
179
significantly more than 50 mg/day equivalent of chlorpromazine will be found to be hyperprolactinemic, yet only a minority will have irregular or absent menses, or galactorrhea. These drugs probably act by limiting the secretion of PIF or by suppressing its action. III.
REGULATION OF MAMMARY GROWTH AND DEVELOPMENT
Our current understanding of the normal sequence of maturational and gestational growth and development of the mammary gland of necessity derives primarily from studies carried out in small laboratory rodents and in dairy animals. Not all the conclusions suggested by these data have been validated in primates. There is general agreement that maturational growth of the mammary gland is probably directly related to prevailing secretory rates of estrogen in vivo. In vitro, the two hormones required for growth and cell division in preparation for prolactin induced milk secretion are insulin and hydrocortisone. In some strains of mice, growth hormone is potently mammotropic (Nandi and Bern, 1961), but human and monkey GH appear to be unique among growth hormones in having lactogenic and mammotropic properties not accountable on the basis of contamination by prolactin. Bovine and ovine GH's are not lactogenic. In organ culture of mouse mammary gland, aldosterone appears to be capable of stimulating ductal differentiation and branching in glands pretreated with insulin and prolactin (Ceriani, 1970), although critical in vivo testing of the need for aldosterone in normal mammogenesis has not been done. No detailed studies of hormonal changes during puberty in experimental animals appear to have been carried out at present. In humans, though in vitro organ culture data are lacking, there is a good deal of information on levels of prolactin and of other potentially relevant hormones at various ages and stages of pubertal and gestational breast development. In addition, a number of hormonally defined disease processes may shed light on the questions at hand. Estrogens stimulate mammary gland hyperplasia in man. Like other estrogen effects, this response is dependent on both the estrogen effects; this response is dependent on both the estrogen level and the duration of tissue exposure to it. Alveolar proliferation, which occurs post-pubertally and is intensified during pregnancy, is highly dependent upon the adequacy of progesterone secretion. It seems reasonable to infer from the organ culture data that normality of cortisol and insulin secretion may be prerequisites for normal mammary growth and development. No systematic clinical data collection on breast development in pubertal girls with Addison's disease has been done, to my knowledge, and juvenile diabetics are treated of course with insulin. Hence, we may have to be satisfied
180
L.S. JACOBS
with guesses and inferences in this matter. It further seems quite likely that euthyroidism may be yet another prerequisite for fully normal breast development. The role played by prolactin in mammogenesis in humans remains unclear. Circulating levels of prolactin are only slightly higher in adult women of reproductive age than in men, and measurements in prepubertal children have generally given results comparable to those in men (Foley et al., 1972; Guyda and Friesen, 1973). In cross-sectional studies of puberty, the observed rise in PRL levels seemed to be gradual (Guyda and Friesen, 1973). Similarly, the rise in estradiol levels appears to be gradual during puberty, with achievement of adult female levels only late in the pubertal process (Ehara, Yen, and Siler, 1975), whereas breast budding and major breast growth in girls are relatively early pubertal events. These data do not support the concept that major changes in estrogen or prolactin levels are responsible for human mammogenesis. Rather, it seems likely that subtle changes in hormonal secretion, coupled with alterations in mammary responsiveness to the hormonal milieu, may be responsible for human pubertal breast growth. It seems reasonable to assume that certain minimal levels of prolactin may be required, in a permissive way, for breast development to occur The occurrence of hypopituitarism or hypogonadism prepubertally is associated with deficient mammary development in the absence of replacement therapy. In hypopituitary girls, treatment with glucocorticoids, thyroid hormone, and growth hormone does not result in notable mammary development. This requires treatment with estrogen, estrogen plus progesterone, or gonadotropins. Prolactin secretion is frequently normal in hypopituitary subjects. In girls with Turner's syndrome, breast development requires sex steroid therapy. Women with isolated growth hormone deficiency develop normal breasts and are fully capable of normal postpartum lactational performance (Tyson, Hwang, Guyda, and Friesen, 1972; Rimoin, Holzman, Merimee, Rabinowitz, Barnes, Tyson, and McKusick, 1968). These considerations strongly suggest that in humans, although eucorticism and euthyroidism may be required for the full expression of mammary growth, maturation, and lactational function, they exert a permissive role only. Growth hormone appears not to play an important role. Since most early breast growth is accounted for by ductal proliferation without much alveolar budding in normal girls whose aldosterone secretion is adequate, one may also question the relevance of the mouse mammary organ culture data cited above to the normal human situation. Although systematic observations in man on the role of androgens in mammary development have not been made, it is clear from
ROLE IN MAMMOGENESIS AND LACTOGENESIS
181
clinical observations in women with virilizing syndromes and in men with alcoholism and gynecomastia that androgens act in opposition to estrogens as regards breast development. No comparison of PRL levels or estradiol levels between buxom and flat-chested women have been made. Table 1 lists the hormones which may be involved in mammary growth and development. The growth and development of the breast during pregnancy is accounted for by the combined effects of hypersecretion of estrogens, progesterone, prolactin, and chorionic somatomammotropin. Considerable alveolar hypertrophy occurs during pregnancy, and if it were not for the inhibitory effect of very high circulating estrogens on the breast response to mammotropic hormones, undoubtedly copious lactation would be a regular feature of the third trimester of pregnancy. As it is, the hormonal bombardment of the breast in late pregnancy is quite remarkable; estradiol and progesterone levels are several hundred-fold higher in late pregnancy than in the follicular phase of the menstrual cycle. Chorionic somatomammotropin levels may rise as high as 10-12 ~g/ml, several orders of magnitude higher than the usual circulating levels of other peptide hormones of comparable molecular weight. Pituitary prolactin levels, under the influence of estrogens, average about 200 ng/ml at term, or roughly twenty times higher than nonpregnant levels. IV.
REGULATION OF MILK PRODUCTION
I will not review here the extensive literature on regulation of lactation in dairy animals, but will rather focus on what is known regarding this subject in women. Many of the same hormonal influences important in the regulation of mammary growth and development play prominent roles in the control of lactation. Failure of lactation in otherwise endocrinologically normal women has been described in association with isolated deficiency of prolactin (Turkington, 1972); these data were obtained by PRL bioassay measurement following administration of phenothiazines, and have yet to be confirmed. From the in vitro organ culture experience, one may suspect that normal insulin secretion, eucorticism, and euthyroidism are important permissive prerequisites for normal lactation. The secretory apparatus must have reached some minimal threshold stage of development under the influence of estrogen and progesterone. Induction of milk protein synthesis is dependent on the mammotropic hormones prolactin or chorionic somatomammotropin. The secretion of milk following all this hormonal preparation further depends on withdrawal of estrogen and progesterone, as well as on intact mechanisms for oxytocin release and myoepithelial basket cell contraction. Tactile stimulation of the breast and nipple results in reflex release of PRL (Kolodny, Jacobs, and Daughaday, 1972; Noel, Suh, and Frantz, 1974) and of oxytocin as well. In endocrinologically normal but nonpregnant women, lactogenesis can be induced by administration of estrogen and a regular program of tac-
L.S. JACOBS
182
tile manipulation of the breast and nipple, followed by abrupt cessation of the estrogen treatment and a waiting period of a few days (Tyson, Khojandi, Ruth, and Andreasson, 1975). Similarly, there is a normal delay of several days following delivery before milk secretion becomes well established postpartum. The important factors responsible for this delay include time required to escape from the inhibition previously exercised by high levels of estrogen and progesterone. With delivery of the placenta, the source of all the placental lactogen (chorionic somatomammotropin) and the great bulk of the progesterone and estrogen produced during late pregnancy is removed. Circulating levels of these hormones fall in accord with their respective half-lives in the serum, but the secretion of prolactin does not change nearly as rapidly. Even in nonbreast-feeding mothers experiencing no suckling stimulus, it may take 2 to 3 weeks for basal prolactin levels to return to nonpregnant levels. This indicates substantial temporal persistence of the estrogen stimulation after its removal, and is consistent with the duration of other biological actions of estrogens. In breast-feeding mothers, the decline of PRL levels is substantially slowed, and nonpregnant fasting levels may not be reached for several weeks. Table 2 summarizes the factors involved in the initiation of postpartum lactation. As has been shown, suckling episodes result in dramatic rises in serum PRL levels during the early puerperium (Tyson et al., 1972). Later during lactation, the serum PRL either increases much less sharply with suckling, or in some cases may not rise measurably at all; basal PRL levels are normal. Although there is some variation in serum PRL from moment to moment, especially in hyperprolactinemic states (Jacobs and Daughaday, 1973), there is reason to believe that randomly obtained levels may nonetheless reflect integrated 24-hour levels reasonably Well (Malarkey, 1975). Thus normal lactation seems to proceed quite well in the absence of accelerated PRL secretion. At the same time, most breast-feeding women remain amenorrheic or oligomenorrheic and rather severely estrogen-deficient. Nonpuerperal galactorrhea is also frequently associated with hypoestrogenism and menstrual irregularities or amenorrhea. Thus, in contrast to breast development, which requires estrogen, lactation seems to require suppression or removal of estrogen. Further, it would appear that Table 2.
FACTORS INVOLVED IN TRE INITIATION OF POSTPARTUM LACTATION Postpartum Time Lag Rapidly Falling Estrogens Rapidly Falling Progesterone Tactile Stimulation of the Breast and Nipple
ROLE IN MAMMOGENESIS AND LACTOGENESIS Table 3.
183
ASSESSMENT OF PROLACTIN EFFECTS ON THE BREAST
Single Determinations VB. Secretory Rates Subtle alterations Maintained over Months or Years Synergistic Effects of Multiple Hormones Variations in Mammary Response to the Hormonal Milieu Systemic VB. Mammary Hormone Concentrations more PRL is required for the initiation of lactation than for its maintenance. Table 3 indicates some of the considerations which need to be kept in mind as one tries to relate measured PRL levels to breast function. The possibility does of course exist that the measured serum PRL level is not an accurate reflection of the hormonal milieu to which the breast is exposed. In order to assess this possibility, and because the presence of immunoreactive PRL in milk had been reported in experimental animals (McMurtry and Malvern, 1974a, 1974b), we have measured the PRL in simultaneously obtained milk and serum specimens in a series of women with nonpuerperal galactorrhea. Figure 1 shows the systematic gradient which exists between these two fluids, the prolactin concentration in milk being significantly higher than that found in serum. Thus the breast does to a certain extent act as a concentrating mechanism. Whether the PRL in milk which is immunologically intact is a more valid index of the hormonal exposure of milk-secreting cells than that in serum remains to be shown, however. Although the mechanism by which apparently intact prolactin finds its say into milk is not known, it seems clear that the hormone must traverse the capillary side of the alveolar cell plasma membrane, the cell interior, a~d the luminal or ductular surface plasma membrane as well. Cell penetrance by polypeptide hormones such as prolactin has in the past been thought not to occur in relation to hormone action; however, recently intracellular polypeptide hormone receptors have been tentatively identified (Posner and Bergeron, 1976) and such a mechanism may be involved in the transfer of prolactin from serum to milk. A similar gradient in PRL seems to exist in postpartum lactating women, so this mechanism is not restricted to situations of abnormal pathophysiology. It is a reasonably frequent occurrence ot find normal PRL levels in the serum of women with nonpuerperal galactorrhea of diverse etiologies. This is especially common in women in whom regular menses persist despite galactorrhea, and no apparent cause for the lactation can be found. Several possible explanations exist for continued galactorrhea despite normal PRL levels. One may postulate that during the early phases of the disorder, higher levels of PRL may have prevailed while lactation was being established.
L.S. JACOBS
184
230 220 140 120
E ...... 0> c: 100 Z I-
u
«
80
-I
0
~
a..
:x: -I
~
60 40 20
o Figure 1.
20
40
60
80
100
120
SERUM PROLACTIN (ng/ml)
Correlation between serum PRL levels and simultaneously obtained milk PRL levels in women with nonpuerperal galactorrhea. The bulk of the data points fall above the diagonal line of identity, indicating a systematic gradient from milk to serum. Whether enhanced entry into, or diminished clearance from milk is the prime factor responsible for the higher milk PRL levels is not known.
ROLE IN MAMMOGENESIS AND LACTOGENESIS Table 4.
185
FACTORS CAPABLE OF INFLUENCING MAMMARY RESPONSIVITY TO PROLACTIN I. II.
III.
PRE-RECEPTOR EVENTS Prior Hormonal Milieu THE PROLACTIN RECEPTOR Affinity Capability POST-RECEPTOR EVENTS Coupling of Hormone Binding to Adeny1ate Cyclase Protein Kinase Phosphorylation of Intracellular Proteins Km, Vmax , of Enzymes Involved in Milk Fat and/or Carbohydrate Synthesis
Alternatively, one may explain such a situation by the suggestion that milk PRL levels may be higher than those in serum, as shown in Figure 1. Finally, an altered, heightened tissue responsivity to PRL could account for such findings. One way in which heightened tissue sensitivity may be explained is by an increase in either the number or the affinity of hormone receptors. Figure 2 demonstrates that just such an increase in PRL affinity does in fact occur in rabbit mammary glands during postpartum lactation. Individual Scatchard plots of PRL to representative early (first postpartum week) and last (just prior to weaning) mammary gland membrane preparations are shown. Although the mechanism responsible for this 2.5-fo1d increase in PRL receptor affinity is not known, such as increase could help account for continued lactation in the face of falling serum PRL levels. Table 4 indicates that changes in hormone receptor affinity represent only one of several possible mechanisms which might underlie enhanced tissue responsivity. On occasion, despite very high serum PRL levels, sometimes in the thousands of nanograms per m1, no mammary manifestations attributable to hyperpro1actinemia may occur. This type of dissociation between clinical manifestations and hormone overproduction is most often seen in patients harboring pituitary adenomas. The absence of gynecomastia in such men, and of galactorrhea in such women, is
L.S. JACOBS
186
0.4
0.3
B/F 0.2
0.1 Early /Pre p #1
I
o
o
1 2 3 456 ng oPRL BOUND I mg PROTEIN I
I
I
I
I
I
52.2 104.4 157 209 261 313 fmoles oPRL BOUND/mg PROTEIN
Figure 2.
Scat chard plots of the interaction of single representative early (first postpartum week) and late (immediately prior to weaning) mammary receptor preps with iodinated ovine prolactin. The binding capacity of the two preparations is similar (60 and 76 fmoles/mg protein) but the affinity of the late prep is 3.8-fold greater than that of the early prep (Ka,l x 1010M-1VS. 2.6 x 109 M- I) .
ROLE IN MAMMOGENESIS AND LACTOGENESIS
187
most satisfactorily explained by postulating that the frequently present hypogonadism results in hypoestrogenism and breasts which are less sensitive to the actions of PRL. Most men with PRL-secreting pituitary adenomas who do have galactorrhea exhibit serum PRL levels which are very markedly elevated, often in excess of 1000 ng/ml. Although many women with such tumors and galactorrhea have PRL levels which are only modestly elevated, suggesting that the clinical expression of hyperprolactinemia at the breast is facilitated by estrogp-ns, nonetheless men with such tumors occasionally have galactorrhea with PRL levels less than 100 ng/ml. Uniformly these men are found to be hypogonadal. Thus it would seem, as in cirrhotics with gynecomastia, that breast responses to sex steroids and to PRL appear to depend on the estrogen/androgen ratio more than the absolute level of either. REFERENCES Ben-Jonathan, N., Oliver, C, and Mical, R.S. (1976). Dopamine secretion into hypophysial portal blood during the estrus cycle and pregnancy in the rat. Program, 58th Meeting, Endocrine Society, Abstract 269, Endocrinology 98, Supplement, p. 191. Birge, C.A., Jacobs, L.S., Hammer, C.T., and Daughaday, W.H. (1970). Catecholamine inhibition of prolactin secretion by isolated rat adenohypophyses. Endocrinology 86, 120-130. Bowers, C.Y., Friesen, H.G., Hwang, P., Guyda, H.J., and Folkers, K. (1971). Prolactin and thyrotropin release in man by synthetic pyroglutamyl-histidyl-prolinamide. Biochem. Biophys. Res. Commun. 45, 1033-1041. Carlson, H.E., Jacobs, L.S., and Daughaday, W.H. (1973). Growth hormone, thyrotropin and prolactin responses to thyrotropinreleasing hormone following diethylstilbestrol pretreatment. J. Clin. Endocrinol. Metab. 37, 488-490. Ceriani, R.L. (1970) Fetal mammary gland differentiation in vitro in response to hormones. I. Morphological findings. Developm. Biol. 21, 506-529. Daughaday, W.H. and Jacobs, L.S. (1972). Normal and pathologic secretion of prolactin in man. In: Endocrinology, Proceedings IV International Congress of Endocrinology (Scow, R.O., ed.), pp. 622-628, Excerpta Medica International Congress Series No. 273. del Pozo, E., Varga, L., Wyss, H., Tolis, G., Friesen, H., Wenner, R., Vetter, L., and Uettwiler, A. (1974). Clinical and hormonal response to bromocriptin (CB-154) in the galactorrhea syndromes. J. Clin. Endocrinol. Metab. 39, 18-36. Ehara, Y., Yen, S.S.C., and Siler, T.M. (1975). Serum prolactin levels during puberty. Amer. J. Obstet. Gynec. 121, 995-997.
L.S. JACOBS
188
Foley, T.P. Jr., Jacobs, L.S., Hoffman, W., Daughaday, W.H., and Blizzard, R.M. (1972). Human prolactin and thyrotropin concentrations in the serums of normal and hypopituitary children before and after the administration of synthetic thyrotropinreleasing hormone. J. Clin. Invest. 51, 2143-2150. Frantz, A.G., Kleinberg, D.L., and Noel, G.L. (1972). Studies on prolactin in man. Rec. Prog. Horm. Res. 28, 527-590. Friesen, H.G., Belanger, C., Guyda, H.J., and Hwang, P. (1972). The synthesis and secretion of placental lactogen and pituitary prolactin. In: Lactogenic HOPmones (Wolstenholme, G.F. W. and Knight, J., eds.), pp. 83-103, Churchill-Livingstone, London. Fuxe, K. (1964). Cellular localization of monoamines in the median eminence and the infundibular stem of some mammals. Z. Zellforsch Mikrosk. Anat. 61, 710-724. Gautvik, K.M., Tashjian, A.H. Jr., Kourides, I.A., Weintraub, B.D., Graeber, C.T., Maloof, F., Suzuki, K., and Zuckerman, J.E. (1974). Thyrotropin-releasing hormone is not the sole physiologic mediator of prolactin release during suckling. New Engl. J. Med. 290, 1162-1165. Hwang, P., Guyda, H., and Friesen, H. (1971). A radioimmunossay for human prolactin. Proc. Natl. Acad. Sci. 68, 1902-1906. Jacobs, L.S. and Daughaday, W.H. (1973). Pathophysiology and control of prol~c.tin sec.retion in patients with pituitary and hypothalamic disease. In: Human Prolactin (Pasteels, J.L. and Robyn, C., eds.), pp. 189, International Congress Series No. 308, Excerpta Medica, Amsterdam. Jacobs, L.S., Mariz, I.K., and Daughaday, W.H. (1972). A mixed heterologous radioimmunoassay for human prolactin. J. Clin. Endocrinol. Metab. 34, 484-490. Jacobs, L.S., Snyder, P.J., Wilber, J.F., Utiger, R.D., and Daughaday, W.H. (1971). Increased serum prolactin after administration of synthetic thyrotropin releasing hormone (TRH) in man. J. Clin. Endocrinol. Metab. 33, 996-999. Kolodny, R.C., Jacobs, L.S., and Daughaday, W.A. (1972). Mammary stimulation causes prolactin secretion in non-lactating women. Nature 238, 284-286. Larson, B.L. and Smith, V.R., eds. (1974). Lactation/A Comprehensive Treatise, 3 volumes, Academic Press. Volume 1: The Mammary Gland/Development and Maintenance, Volume 2: Biosynthesis and Secretion of Milk/Diseases, and Volume 3: Nutrition and
Biochemistry of Milk/Maintenance of Lactation.
Lu, K.H. and Meites, J. (1973). Effects of serotonin precursors and melatonin on serum prolactin release in rats. Endocrinology 93, 152-155. MachI in , L.J., Jacobs, L.S., Cirulis, N., Kimes, R., and Miller, R. (1974). An assay for growth hormone and prolactin-releasing activities using a bovine pituitary cell culture system. Endocrinology 95, 1350-1358.
ROLE IN MAMMOGENESIS AND LACTOGENESIS
189
Maclndoe, J.H. and Turkington, R.W. (1973). Stimulation of human prolactin secretion by intravenous infusion of L-tryptophan. J. Clin. Invest. 52, 1972-1978. Macleod, R.M. and Lelli~eyer, J.E. (1974). Studies on the mechanisms of the dopamine-mediated inhibition of prolactin secretion. Endocrinology 94, 1077-1085. Malarkey, W.B. (1975). Nonpuerperal lactation and normal prolactin regulation. J. Clin. Endocrinol. Metab. 40, 198-204. McMurtry, J.P. and Malven, P.V. (1974a). Radioimmunoassay of endogenous and exogenous prolactin in milk of rats. J. Endocrinol. 61, 211-217. McMurtry, J.P. and Malven, P.V. (1974b). Experimental alterations of prolactin levels in goat milk and blood plasma. Endocrinology 95, 559-564. Meites, J. and Nicoll, C.S. (1966). Adenohypophysis: Prolactin. Ann. Rev. Physiol. 28, 57-88. Nandi, S. and Bern, H.A. (1961). The hormone responsible for lactogenesis in BALB-cCrgl mice. Gen. Compo Endocrinol. 1, 195210. Nicoll, C.S., Fiorindo, R.P., McKennee, C.R., and Parsons, J.A. (1970). Assay of hypothalamic factors which regulate prolactin secretion. In: Hypophysiotropic Hormones of the Hypothalamus: Assay and Chemistry (Meites, J., ed.), pp. 115-144, Williams and Wilkins, Baltimore. Noel, G.L., Suh, H.K., and Frantz, A.G. (1974). Prolactin release during nursing and breast stimulation in postpartum and nonpostpartum subjects. J. Clin. Endocrinol. Metab. 38, 413-423. Posner, B.I. and Bergeron, J.J.M. (1976). Intracellular polypeptide hormone receptors. Program, 58th Meeting, Endocrine Society, Abstract 218, Endocrinology 98, Supplement, pp. 165. Rimoin, D.L., Holzman, G.B., Merimee, T.J., Rabinowitz, D., Barnes, A.C., Tyson, J.E., and McKusick, V.A. (1968). Lactation in the absence of human growth hormone. J. Clin. Endocrinol. Metab. 28, 1183-1188. Shaar, C.J. and Clemens, J.A. (1974). The role of catecholamines in the release of anterior pituitary prolactin in vitro. Endocrinology 95, 1201-1212. Snyder, P.J., Jacobs, L.S., Utiger, R.D., and Daughaday, W.H. (1973). Thyroid hormone inhibition of the prolactin response to thyrotropin-releasing hormone. J. Clin. Invest. 52, 2324-2329. Snyder, P.J., Jacobs, L.S., Rabello, M.M., Sterling, F.H., Shore, R.N., Utiger, R.D., and Daughaday, W.H. (1974). Diagnostic value of thyrotropin-releasing hormone in pituitary and hypothalamic diseases: Assessment of thyrLtropin and prolactin secretion in 100 patients. Annals Int. Med. 81, 751-757. Turkington, R.W. (1972). Phenothiazine stimulation test for prolactin reserve: The syndrome of isolated prolactin deficiency. J. Clin. Endocrinol. Metab. 34, 247-250.
L.S. JACOBS
190
Tyson, J.E., Hwang, P., Guyda, H., and Friesen, H.G. (1972). Studies of prolactin secretion in human pregnancy. Amer. J. Obstet. Gynecol. 113, 14-20. Tyson, J.E., Khojandi, M., Huth, J., and Andreasson, B. (1975). The influence of prolactin secretion on human lactation. J. Clin. Endocrinol. Metab. 40, 764-773. Valverde, C., and Chieffo, V. (1971). Prolactin releasing factors in porcine hypothalamic extracts. Program, 53rd Meeting, Endocrine Society, Abstract 83, Endocrinology 88, Supplement, p. A84. Woolf, P.D., Jacobs, L.S., Donofrio, R., Burday, S.Z., and Schalch, D.S. (1974). Secondary hypopituitarism: Evidence for continuing regulation of hormone release. J. Clin. Endocrinol. Metab. 38, 71-76. Yen, S.S.C., Ehara, Y., and Siler, T.M. (1974). Augmentation of prolactin secretion by estrogen in hypogonadal women. J. Clin. Invest. 53, 652-655. DISCUSSION AFTER DR. JACOBS' PAPER Dr. Barna1.Je U I wUlulel- if you know af any measurement R of prolact in levels
during the menstrual cycle. Dr. Jacobs
Yes, several groups of investigators have measured prolactin levels in a very large number of normally cycling women. Although one group of European investigators claim that there is a significant increase in prolactin during the luteal phase, I think by and large most investigators have found no significant differences, so I would say that there is no significant variation in prolactin during the menstrual cycle. Dr. Barna1.Je U
Except for the possibility that nocturnal levels might be higher.
Dr. Jacobs
Yes, that remains a possibility.
Dr. Klein
Dr. Yen published data which showed that the nocturnal rise in prolactin at approximately mid-cycle is about twice as high as during either the follicular or luteal phase, but that's on only one patient (Ehara, Siler, Vandenberg, Sinha, and Yen, 1973). Has anyone investigated the biological activity of the prolactin found in some of the patients with tumors who don't lactate? Could this be immunoassayable prolactin that is not biologically active?
ROLE IN MAMMOGENESIS AND LACTOGENESIS
191
Dr. Jacobs
Some of the best data on this question were shown by Dr. Frantz in terms of the good general correlation between bioassay and radioimmunoassay results. In addition, Dr. Friesen has shown excellent correlation between radioimmunoassay and radioreceptor assay results in a large number of specimens spanning both the physiologic and pharmacologic ranges (Friesen, 1973). These data suggest strongly that the failure to observe lactation in many women with prolactinproducing pituitary adenomas is the result of variable breast responses rather than biologically inactive hormone.
Dr. Sage
I'm not familiar with the clinical literature, but the La Leche league reports on a number of cases of women who have never had children, but who adopt children and are capable of producing milk. Have any of these been followed clinically and the prolactin levels assayed?
Dr. Jacobs
Not to my knowledge. It is well documented, however, that postmenopausal grandmothers have successfully suckled their grandchildren. This used to be common in certain African tribal communities in which the men hunted and the reproductive age women did all the agricultural field work. The critical factor here, and in the La Leche reports to which you refer, is probably tactile stimulation of the breast and especially the nipple. Such tactile stimulation can, reflexly, cause release of prolactin. The reflex appears to be estrogen-sensitive. I might add a fascinating bit of what one might call "mammaryology" in the marsupial. Kangaroos not infrequently give birth while still breastfeeding, and it is not unusual, I understand, for the rather advanced joey at age one or more to be breastfeeding at the same time that there is a newborn in the pouch breastfeeding. It has been shown that the composition of the milk produced by the two teats is markedly different. There are differences in protein electrophoretic patterns and differences in fat composition. Since they corne from the same mother, one would presume that the hormonal milieu is the same. Hence, the observed differences in composition may relate to differences in quantity, quality, or frequency of the tactile stimulation produced by the neonate and that of the joey. DISCUSSION REFERENCES Ehara, Y., Siler, T., Vandenberg, G., Sinha, Y.M., and Yen, S.S.C. (1973). Circulating prolactin levels during the menstrual cycle: Episodic release and diurnal variation. Amer. J. Obstet. Gynec. 117, 962-970. Friesen, H.G. (1973). In Human Prolactin (Pasteels, J.L. and Robyn, C., eds.), Ies #308, Excerpta Medica, Amsterdam.
